A large percentage of mutations in genetic disease and cancer are CAET transitions at CpG sites, generated mainly by deamination of 5-methylcytosine (m5C) to give G.T mispairs. Cytosine methylation at CpG sites is a mark for transcriptional silencing that is central to many cellular processes and essential for embryogenesis. Thus, maintaining CpG integrity is important for mutation avoidance and proper transcriptional regulation. Two DNA glycosylases recognize G.T lesions, thymine DNA glycosylase (TDG) and methyl binding domain IV (MBD4). Initiating the base excision repair (BER) pathway, these glycosylases flip the target nucleotide (dT) into their active site and cleave the base-sugar (N-glycosylic) bond, producing an abasic (AP) site. Repair continues with AP endonuclease (APE1) and downstream BER enzymes. TDG and MBD4 face the daunting task of removing a normal base from a mismatched pair, and must balance the needs for efficient G.T repair and avoidance of undamaged DNA, which may limit their activity. Given the biological need to maintain CpG integrity for mutation avoidance and transcriptional regulation, it is important to obtain a detailed understanding of how TDG and MBD4 recognize and remove lesions, why their activity is slow for G.T mispairs (which may impact CpG mutability) and how their activity is regulated by APE1. Towards this end, we propose a powerful combination of biochemical, biophysical, and structural methods to achieve four specific aims: (i) We will use transient kinetics and equilibrium binding methods to determine the parameters that govern lesion recognition, lesion excision, and product release for TDG and MBD4, revealing why their turnover is slow for G.T lesions and fast for excision of 5-halogenated uracils such as 5FU. (ii) We will determine the crystal structure of TDG (catalytic domain) bound to DNA containing a non-cleavable substrate analog, revealing interactions that promote G.T specificity. To understand how TDG interrogates but does not act upon CpG sites, we will attempt to solve the structure of TDG bound to CpG DNA. (iii) We will use transient kinetics, and NMR methods with a stable TDG-AP-DNA complex, to reveal how APE1 regulates TDG activity (enhances its turnover). We will also test the hypothesis that SUMOylation of TDG is required for timely product release and efficient repair of G.T lesions. (iv) We will use NMR to determine how the intrinsically disordered N-terminal region of TDG enables efficient G.T repair, and how the disordered C-terminal region forms non-covalent interactions with SUMO proteins, which is important for SUMO regulation of TDG activity and TDG binding to SUMO-modified proteins (i.e., p731, PML). Understanding the function of intrinsically disordered protein regions is a major challenge in structural biology, and our studies will contribute to this emerging field. Successful completion of these studies will advance our understanding of how mutagenic G.T lesions are recognized and repaired in humans, why CpG sites are mutational hotspots, and how TDG mediates the cytotoxicity of 5FU, a widely used anti-cancer drug, with implications for the role of TDG and MBD4 in cancer, genetic disease, and 5FU chemotherapy.